Customizable Scale-Out Key-Value Stores

Anwar, Ali; Cheng, Yue; Huang, Hai; Han, Jingoo; Simh@ornl.gov, Hyogi; Lee, Dongyoon; Douglis, Fred; Butt, Ali R.

doi:10.1109/TPDS.2020.2982640

Title: Customizable Scale-Out Key-Value Stores

Full Record
Other Related Research

Abstract

Enterprise KV stores are often not well suited for HPC applications, and thus cumbersome end-to-end KV design customization is required to meet the needs of modern HPC applications. To this end, in this article we present bespoKV, an adaptive, extensible, and scale-out KV store framework. bespoKV decouples the KV store design into the control plane for distributed management and the data plane for local data store. For the control plane, bespoKVprovides pre-built modules, called controlets, supporting common distributed functionalities (e.g., replication, consistency, and topology) and their various combinations. This decoupling allows bespoKV to take a user-provided single-server KV store, called a datalet, and transparently enables a scalable and fault-tolerant distributed KV store service. The resulting distributed stores are also adaptive to consistency or topology requirement changes and can be easily extended for new types of services. Such specializations enable innovative uses of KV stores in HPC applications, especially for emerging applications that utilize KV-friendly workloads. We evaluate bespoKV in a local testbed as well as in a public cloud settings. Experiments show that bespoKV-enabled distributed KV stores scale horizontally to a large number of nodes, and performs comparably and sometimes 1.2× to 2.6× better than the state-of-the-art systems.

Authors:

^[1];

^[2]; Huang, Hai ^[3]; Han, Jingoo ^[4];

^[5];

^[6];

^[7]; Butt, Ali R. ^[4]

IBM Research-Almaden, San Jose, CA (United States)
George Mason Univ., Fairfax, VA (United States)
IBM Research, Ossining, NY (United States)
Virginia Polytechnic Inst. and State Univ. (Virginia Tech), Blacksburg, VA (United States)
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Stony Brook Univ., NY (United States)
Perspecta Labs, Basking Ridge, NJ (United States)

Publication Date:: 2020-03-30

Research Org.:: Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)

Sponsoring Org.:: USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)

OSTI Identifier:: 1632067

Grant/Contract Number:: AC05-00OR22725

Resource Type:: Accepted Manuscript

Journal Name:: IEEE Transactions on Parallel and Distributed Systems

Additional Journal Information:: Journal Volume: 31; Journal Issue: 9; Journal ID: ISSN 1045-9219

Publisher:: IEEE

Country of Publication:: United States

Language:: English

Subject:: 97 MATHEMATICS AND COMPUTING

Citation Formats


                    Anwar, Ali, Cheng, Yue, Huang, Hai, Han, Jingoo, Simh@ornl.gov, Hyogi, Lee, Dongyoon, Douglis, Fred, and Butt, Ali R. Customizable Scale-Out Key-Value Stores.  United States: N. p., 2020. 
Web.  doi:10.1109/TPDS.2020.2982640.

Copy to clipboard


                    Anwar, Ali, Cheng, Yue, Huang, Hai, Han, Jingoo, Simh@ornl.gov, Hyogi, Lee, Dongyoon, Douglis, Fred, & Butt, Ali R. Customizable Scale-Out Key-Value Stores.  United States.  https://doi.org/10.1109/TPDS.2020.2982640

Copy to clipboard


                    Anwar, Ali, Cheng, Yue, Huang, Hai, Han, Jingoo, Simh@ornl.gov, Hyogi, Lee, Dongyoon, Douglis, Fred, and Butt, Ali R. Mon .  
"Customizable Scale-Out Key-Value Stores".  United States.  https://doi.org/10.1109/TPDS.2020.2982640.  https://www.osti.gov/servlets/purl/1632067.

Copy to clipboard


                    
@article{osti_1632067,

  title        = {Customizable Scale-Out Key-Value Stores},

  author       = {Anwar, Ali and Cheng, Yue and Huang, Hai and Han, Jingoo and Simh@ornl.gov, Hyogi and Lee, Dongyoon and Douglis, Fred and Butt, Ali R.},

  abstractNote = {Enterprise KV stores are often not well suited for HPC applications, and thus cumbersome end-to-end KV design customization is required to meet the needs of modern HPC applications. To this end, in this article we present bespoKV, an adaptive, extensible, and scale-out KV store framework. bespoKV decouples the KV store design into the control plane for distributed management and the data plane for local data store. For the control plane, bespoKVprovides pre-built modules, called controlets, supporting common distributed functionalities (e.g., replication, consistency, and topology) and their various combinations. This decoupling allows bespoKV to take a user-provided single-server KV store, called a datalet, and transparently enables a scalable and fault-tolerant distributed KV store service. The resulting distributed stores are also adaptive to consistency or topology requirement changes and can be easily extended for new types of services. Such specializations enable innovative uses of KV stores in HPC applications, especially for emerging applications that utilize KV-friendly workloads. We evaluate bespoKV in a local testbed as well as in a public cloud settings. Experiments show that bespoKV-enabled distributed KV stores scale horizontally to a large number of nodes, and performs comparably and sometimes 1.2× to 2.6× better than the state-of-the-art systems.},

  doi          = {10.1109/TPDS.2020.2982640},

  journal      = {IEEE Transactions on Parallel and Distributed Systems},

  number       = 9,

  volume       = 31,

  place        = {United States},

  year         = {2020},

  month        = {3}

}

Copy to clipboard

Journal Article:

Free Publicly Available Full Text

Accepted Manuscript (DOE)

Publisher's Version of Record

https://doi.org/10.1109/TPDS.2020.2982640

Other availability

Search WorldCat to find libraries that may hold this journal

Citation Metrics:

Cited by: 4 works

Citation information provided by
Web of Science

Save / Share:

Export Metadata

Save to My Library

Similar Records in DOE PAGES and OSTI.GOV collections:

BESPOKV: Application Tailored Scale-Out Key-Value Stores

Conference Anwar, Ali ; Cheng, Yue ; Huang, Hai ; ...

Enterprise KV stores are not well suited for HPC applications, and entail customization and cumbersome end-to-end KV design to extract the HPC application needs. To this end, in this paper we present BESPOKV, an adaptive, extensible, and scale-out KV store framework. BESPOKV decouples the KV store design into the control plane for distributed management and the data plane for local data store. BESPOKV takes as input a single-server KV store, called a datalet, and transparently enables a scalable and fault-tolerant distributed KV store service. The resulting distributed stores are also adaptive to consistency or topology requirement changes and can bemore »« less
https://doi.org/10.1109/SC.2018.00005

Full Text Available
Rethinking key–value store for parallel I/O optimization

Journal Article Kougkas, Anthony ; Eslami, Hassan ; Sun, Xian-He ; ... - International Journal of High Performance Computing Applications

Key-value stores are being widely used as the storage system for large-scale internet services and cloud storage systems. However, they are rarely used in HPC systems, where parallel file systems are the dominant storage solution. In this study, we examine the architecture differences and performance characteristics of parallel file systems and key-value stores. We propose using key-value stores to optimize overall Input/Output (I/O) performance, especially for workloads that parallel file systems cannot handle well, such as the cases with intense data synchronization or heavy metadata operations. We conducted experiments with several synthetic benchmarks, an I/O benchmark, and a real application.more »« less
https://doi.org/10.1177/1094342016677084
Distributed Logical Analytic Domains (DLADs) for Distributed HPC Security

Technical Report Ros-Giralt, Jordi

The DOE estimates that the costs of a cyber attack onto its main high performance computing (HPC) facilities could potentially be in the order of billions of US dollars. The same conclusion is reached in the commercial sector, with a recent study by Oxford Economics reporting a combined annualized loss of $12.5 billion in shareholder value for publicly traded companies (more than $50 billion aggregated during the past four years) due to cybersecurity breaches. On average, the cost of cybercrime for organizations is worsening, increasing at an annualized compound rate of 23 percent, according to Accenture’s 2017 report on themore »« less
iez: Resource Contention Aware Load Balancing for Large-Scale Parallel File Systems

Conference Wadhwa, Bharti ; Paul, Arnab K. ; Neuwirth, Sarah ; ...

Parallel I/O performance is crucial to sustaining scientific applications on large-scale High-Performance Computing (HPC) systems. However, I/O load imbalance in the underlying distributed and shared storage systems can significantly reduce overall application performance. There are two conflicting challenges to mitigate this load imbalance: (i) optimizing systemwide data placement to maximize the bandwidth advantages of distributed storage servers, i.e., allocating I/O resources efficiently across applications and job runs; and (ii) optimizing client-centric data movement to minimize I/O load request latency between clients and servers, i.e., allocating I/O resources efficiently in service to a single application and job run. Moreover, existing approachesmore »« less
https://doi.org/10.1109/IPDPS.2019.00070

Full Text Available
PapyrusKV: A High-Performance Parallel Key-Value Store for Distributed NVM Architectures

Conference Kim, Jungwon ; Lee, Seyong ; Vetter, Jeffrey S.

This paper introduces PapyrusKV, a parallel embedded key-value store (KVS) for distributed high-performance computing (HPC) architectures that offer potentially massive pools of nonvolatile memory (NVM). PapyrusKV stores keys with their values in arbitrary byte arrays across multiple NVMs in a distributed system. PapyrusKV provides standard KVS operations such as put, get, and delete. More importantly, PapyrusKV provides advanced features for HPC such as dynamic consistency control, zero-copy workflow, and asynchronous checkpoint/restart. Beyond filesystems, PapyrusKV provides HPC programmers with a high-level interface to exploit distributed NVM in the system, and it transparently organizes data to achieve high performance. Also, it allowsmore »« less
https://doi.org/10.1145/3126908.3126943

Similar Records